[FIX] html_sanitize: handle the case of empty documents (after sanitization).

[IMP] test_mail: do not mute logger

bzr revid: chs@openerp.com-20131113113715-tzaog0d6o2u4q0x4

openerp/tests/test_mail.py

@@ -23,7 +23,7 @@
 ##############################################################################
 
 import unittest2
-from openerp.tools import html_sanitize, html_email_clean, append_content_to_html, plaintext2html, mute_logger
+from openerp.tools import html_sanitize, html_email_clean, append_content_to_html, plaintext2html
 
 HTML_SOURCE = """
 <font size="2" style="color: rgb(31, 31, 31); font-family: monospace; font-variant: normal; line-height: normal; ">test1</font>
@@ -178,7 +178,6 @@ class TestSanitizer(unittest2.TestCase):
             html = html_sanitize(content)
             self.assertEqual(html, expected, 'html_sanitize is broken')
 
-    @mute_logger('openerp.tools.mail')
     def test_evil_malicious_code(self):
         # taken from https://www.owasp.org/index.php/XSS_Filter_Evasion_Cheat_Sheet#Tests
         cases = [

openerp/tools/mail.py

@@ -2,7 +2,7 @@
 ##############################################################################
 #
 #    OpenERP, Open Source Business Applications
-#    Copyright (C) 2012 OpenERP S.A. (<http://openerp.com>).
+#    Copyright (C) 2012-2013 OpenERP S.A. (<http://openerp.com>).
 #
 #    This program is free software: you can redistribute it and/or modify
 #    it under the terms of the GNU Affero General Public License as
@@ -52,17 +52,19 @@ def html_sanitize(src):
     # html encode email tags
     part = re.compile(r"(<(([^a<>]|a[^<>\s])[^<>]*)@[^<>]+>)", re.IGNORECASE | re.DOTALL)
     src = part.sub(lambda m: cgi.escape(m.group(1)), src)
-    
+
     # some corner cases make the parser crash (such as <SCRIPT/XSS SRC=\"http://ha.ckers.org/xss.js\"></SCRIPT> in test_mail)
     try:
         cleaner = clean.Cleaner(page_structure=True, style=False, safe_attrs_only=False, forms=False, kill_tags=tags_to_kill, remove_tags=tags_to_remove)
         cleaned = cleaner.clean_html(src)
-    except TypeError, e:
+    except TypeError:
         # lxml.clean version < 2.3.1 does not have a kill_tags attribute
         # to remove in 2014
-        cleaner = clean.Cleaner(page_structure=True, style=False, safe_attrs_only=False, forms=False, remove_tags=tags_to_kill+tags_to_remove)
+        cleaner = clean.Cleaner(page_structure=True, style=False, safe_attrs_only=False, forms=False, remove_tags=tags_to_kill + tags_to_remove)
         cleaned = cleaner.clean_html(src)
-    except:
+    except Exception, e:
+        if isinstance(e, etree.ParserError) and 'empty' in str(e):
+            return ""
         _logger.warning('html_sanitize failed to parse %s' % (src))
         cleaned = '<p>Impossible to parse</p>'
     return cleaned